zero-shot learning

From Wiktionary, the free dictionary
Jump to navigation Jump to search

English

[edit]
English Wikipedia has an article on:
Wikipedia

Noun

[edit]

zero-shot learning (uncountable)

  1. (machine learning) A problem setup in machine learning, where, at test time, a learner observes samples from classes that were not observed during training and attempts to predict the class that they belong to, typically based on auxiliary information.
    Coordinate terms: few-shot learning, N-shot learning, one-shot learning